A Novel Approach of Hybrid Method of Machine Learning and Statistical Properties in Pattern Recognition

In this article, we present new method of hybrid intelligent systems. Intelligent systems are very useful for prediction in any situation and are a new wave of embedded and real-time systems that are highly connected, with massive processing power and performing complex process. Here, we present a method of hybrid systems for prediction of the roughness of robot laser hardened specimens, if we cannot measure or calculate one of parameters in existing models of intelligent systems. Also, we have a model of intelligent systems for prediction of the roughness of hardened specimens, but we cannot calculate one parameter in this model. Thus, we develop a new method of hybrid intelligent systems to solve this problem. We develop a hybrid of genetic programming, neural network and multiple regression. In this article, we present new method for statistical pattern recognition. Statistical pattern recognition relates to the use of statistical techniques for analyzing data measurements in order to extract information and make justified decisions. This solution to the presented problem has applications for metallurgy and mechanical engineering. With the hybrid system intelligent system, we increase production of the process of laser hardening, because we decrease the time of the process and increase the topographical property of the materials. This paper explores the use of an intelligent system with a new hybrid method of intelligent systems to predict the mechanical properties of robot laser hardened specimens, if we do not know one parameter of the existing model.